At the heart of the review is the idea that host seeking unfolds through overlapping phases: activation, orientation, approach, landing, probing, and feeding. Each phase depends on different cues with different reliability. Carbon dioxide exhaled by a vertebrate is a long-range signal that can activate a hungry female from tens of meters away, but it is not host-specific, since any breathing animal produces it. Skin odors carry the identity of a particular host but operate at shorter ranges. Heat and humidity become informative only at close quarters, and visual cues dominate in bright conditions. Because no single cue suffices across the entire journey, mosquitoes must stitch together fragments of information, and the authors emphasize that internal states such as hunger, mating status, and egg-development stage, along with learned associations, bridge the temporal gaps between cues encountered at different moments.

To explain how this stitching happens, the review outlines three sensory architectures that can contribute to host recognition. The first is feedforward multisensory integration, in which signals from different modalities converge in higher brain centers such as the antennal lobe, the lateral horn, and the mushroom body, producing a combined representation of the host. The second is cross-modal gating, in which one sensory channel modulates the sensitivity of another; the classic example is the way carbon dioxide primes the olfactory system to respond more strongly to skin odors. The third, and arguably the most conceptually important, is closed sensorimotor feedback: a mosquito’s own movements change the sensory input it encounters next, so flying upwind toward a plume, casting sideways when the plume is lost, and steering during landing are all acts of active sensing. The insect is not a passive receiver of stimuli but an agent that structures its own perceptual world through action.

This active-sensing perspective carries a technical implication that the authors highlight: host seeking cannot be fully understood by presenting mosquitoes with fixed stimuli in a wind tunnel and recording their choices. Behavior unfolds over seconds to minutes within a single host-seeking bout, and the trajectory of the insect determines which cues it samples and in what order. Sensorimotor dynamics therefore create a loop in which perception guides movement and movement reshapes perception. The review argues that future experiments should be designed to probe this loop directly, for example by tracking how mosquitoes modulate their flight patterns to resolve ambiguous plumes or how they adjust probing behavior on the skin in response to the thermal and chemical feedback they receive from each attempt.

The synthesis also examines how infection changes the picture. Pathogens such as dengue virus, Zika virus, and West Nile virus can modify the host-derived sensory signals that a mosquito encounters, and they can also alter the mosquito’s own responsiveness, probing behavior, feeding persistence, locomotion, and neuromodulation. The authors are careful to grade the evidence. Studies showing that infected hosts or infected mosquitoes differ in relevant traits provide association. Experiments that causally manipulate the pathogen or the mosquito’s physiology establish modulation. Only in a smaller subset of systems do the observed patterns fit what would be expected of adaptive manipulation, in which the pathogen benefits specifically from increased transmission. This tiered framing matters because claims of manipulation are easy to overstate, and the review provides a vocabulary for distinguishing strong from weak inference.

One of the most consequential findings the review consolidates is that infection can reshape the odor landscape itself. Work on dengue and Zika, for instance, has shown that infection can change the volatile chemicals emitted by host skin, making infected individuals more attractive to mosquitoes in ways that plausibly enhance transmission. Conversely, the mosquito’s own infection status can alter how its nervous system processes those cues. Because the extrinsic incubation period, the time a pathogen needs to become transmissible, must align with the mosquito’s blood-feeding schedule, even modest infection-associated shifts in host-seeking or feeding behavior can have disproportionate effects on epidemiological outcomes. The review stresses that these behavioral changes occur on timescales of seconds to minutes within a feeding bout, while transmission emerges over much longer timescales, and connecting the two levels remains an open challenge.

Redundancy emerges as a recurring theme with direct practical consequences. Because host recognition is multimodal, knocking out a single sensory pathway often fails to abolish host seeking. Mosquitoes with impaired carbon dioxide detection can still locate hosts using skin odors and heat; disrupting one family of odorant receptors does not silence the ionotropic receptor channel or the trigeminal-like pathways that detect thermal and humidity cues. This sensory compensation explains why many laboratory interventions, including repellents such as DEET and IR3535 and genetic manipulations of receptor co-receptors such as Orco, show reduced or context-dependent performance in the field. The review argues that intervention studies should evaluate whole-animal phenotypes and mosquito-human contact rates, not just responses at the receptor level, because the intact animal can route around a blocked pathway.

Ecological context is the second major caveat the authors raise for control strategies. Field studies consistently show that the effectiveness of attractant-baited traps, spatial repellents, and odor-based interventions depends on the local environment, the composition of competing host odors, wind conditions, and the species and physiological state of the local mosquito population. A lure that outperforms a human in a semi-field enclosure may fail in a village where natural host cues are abundant and varied. The review therefore proposes a tiered evaluation framework spanning receptor-level mechanisms, whole-animal behavior, effects on mosquito-human contact, and ultimately transmission or disease endpoints, and it warns that laboratory results should not be extrapolated to field performance without explicit testing across contexts.

The microbiome adds yet another layer of complexity. Skin microbiota shape the volatile profile that makes one human more attractive to mosquitoes than another, and the mosquito’s own microbial community can influence its olfactory sensitivity and feeding behavior. The review integrates this evidence into the multimodal framework, suggesting that microbiome-mediated variation in host odors and vector competence represents a modifiable component of the transmission cycle. Combined with the growing recognition that learning allows mosquitoes to adjust their host preferences based on experience, the picture that emerges is of a vector whose host-seeking behavior is plastic at multiple levels: neural, microbial, and experiential.

The review closes by generating testable predictions for active sensing, sensory compensation, infection-associated modulation, and the persistence of intervention effects across laboratory and field settings, and it identifies unresolved sites of neural convergence in the mosquito brain as priorities for future research. For a field in which mosquito-borne diseases such as malaria, dengue, and West Nile fever continue to impose enormous burdens, the message is clear: durable transmission control will come not from blocking a single cue but from understanding the full sensorimotor loop that connects a flying insect to its next blood meal. By mapping that loop, from plume-following flight to the final probing of the skin, and by specifying where pathogens intervene within it, the authors offer both a conceptual framework and a practical roadmap for the next generation of vector-control research.

Subject of Research: Multimodal sensory integration and active sensing in mosquito host recognition and its implications for pathogen transmission control

Article Title: Multimodal host recognition in mosquitoes: sensory integration, active sensing, and transmission control

Article References: Liu, Z., Shu, Z., Bai, Y., Zhang, T., Shi, H., Zhang, D., Liu, G., & Jin, Y. (2026). Multimodal host recognition in mosquitoes: sensory integration, active sensing, and transmission control. Parasites & Vectors. https://doi.org/10.1186/s13071-026-07702-9

Image Credits: AI Generated

DOI: 10.1186/s13071-026-07702-9

Keywords: mosquitoes, host seeking, multisensory integration, active sensing, carbon dioxide, odorant receptors, dengue virus, Zika virus, West Nile virus, pathogen-vector interactions, vector control, sensory neurobiology